A5049863799- Cooperative Communication and Network Coding
- Data Management and Algorithms
- Privacy-Preserving Technologies in Data
- Advanced Clustering Algorithms Research
- Wireless Communication Security Techniques
- Quantum Computing Algorithms and Architecture
- Quantum Mechanics and Applications
- Quantum Information and Cryptography
- Big Data and Business Intelligence
- Advancements in Semiconductor Devices and Circuit Design
- Face and Expression Recognition
- Complex Network Analysis Techniques
University of Copenhagen
2024
Correlation Clustering is a classic clustering objective arising in numerous machine learning and data mining applications. Given graph $G=(V,E)$, the goal to partition vertex set into clusters so as minimize number of edges between plus missing within clusters. The problem APX-hard best known polynomial time approximation factor 1.73 by Cohen-Addad, Lee, Li, Newman [FOCS'23]. They use an LP with $|V|^{1/\epsilon^{\Theta(1)}}$ variables for some small $\epsilon$. However, due practical...
Correlation Clustering is a classic clustering objective arising in numerous machine learning and data mining applications. Given graph G=(V,E), the goal to partition vertex set into clusters so as minimize number of edges between plus missing within clusters. The problem APX-hard best known polynomial time approximation factor 1.73 by Cohen-Addad, Lee, Li, Newman [FOCS'23]. They use an LP with |V|1/єΘ(1) variables for some small є. However, due practical relevance correlation clustering,...
This paper investigates the impact of noise in quantum query model, a fundamental framework for algorithms. We focus on scenario where oracle is subject to non-unitary (or irreversible) noise, specifically under \textit{faulty oracle} fails with constant probability and acts as identity. Regev Schiff (ICALP'08) showed that advantage lost search problem this model. Our main result shows every algorithm can be made robust model roughly quadratic blow-up complexity, thereby preserving speedup...
Integer data is typically made differentially private by adding noise from a Discrete Laplace (or Gaussian) distribution. We study the setting where differential privacy of counting query achieved using bit-wise randomized response, i.e., independent, random bit flips on encoding answer. Binary error-correcting codes transmitted through noisy channels with independent are well-studied in information theory. However, such unsuitable for since they have (by design) high sensitivity,...
In this work we study the phenomenon of self-testing from first principles, aiming to place versatile concept on a rigorous mathematical footing. Self-testing allows classical verifier infer quantum mechanical description untrusted devices that she interacts with in black-box manner. Somewhat contrary paradigm, existing results tend presuppose conditions constrain operation devices. A common assumption is these perform projective measurement pure state. Naturally, absence any prior knowledge...